Monocular motion stereo-based free parking space detection apparatus and method

Abstract

A monocular motion stereo-based automatic free parking space detection system is disclosed. The system acquires image sequences with a single rearview fisheye camera, three-dimensionally reconstructs the vehicle rearview by using point correspondences, and recovers metric information from a known camera height to estimate the positions of adjacent vehicles thereby detecting the free parking spaces. By using de-rotation-based feature selection and 3D structure mosaicking the degradation of the 3D structure near the epipole is solved and it is not necessary to use the unreliable odometry due to its accuracy depending on road conditions.

Description

CROSS-REFERENCE TO RELATED APPLICATION[0001]This non-provisional application claims priority under 35 U.S.C §119(a) on Patent Application No. 10-2008-0028581 filed in Korea on Mar. 27, 2008, the entire contents of which are hereby incorporated by reference.TECHNICAL FIELD[0002]The present invention relates to a free parking detection apparatus and method. More particularly, the present invention relates to a monocular motion stereo-based free parking space detection apparatus and method for capturing the automobile rearview image and detecting from the image a free parking space.BACKGROUND OF THE DISCLOSURE[0003]An automatic parking system provides convenience for drivers by automatically finding free parking spaces and steering their automobiles toward them. Recently, there have been increased interests in automatic parking systems. By the customers' interests and the success stories of several automatic parking systems, many car manufacturers and component manufacturers are prepar...

AI Technical Summary

Benefits of technology

[0014]Yet another embodiment of the present disclosure provides a method for a vehicle free parking space detection apparatus to detect the free parking space comprising the steps of tracking feature points by receiving input images from an image captor and acquiring point correspondences through analyzing the input images; generating a three-dimensional structure through three-dimensionally reconstructing the free parking space by using the feature points; reconstructing the three-dimensional structure through reselecting the feature points by using a de-rotation-based feature selection; generating a three-dimensional structure mosaic through mosaicking the reconstructed three-dimensional structure by estimating a similarity transformation; acquiring metric information from the three-dimensional structure mosaic by using an image captor height ratio; and detecting the free parking space by estimating the positions of adjacent vehicles in the three-dimensional structure mosaic. As described above, the present disclosure can solve the degradation of three-dimensional structures near the epipole by using de-rotation-based feature selection and three-dimensional structure mosaicking.

[0015]In addition, the present disclosure suggests an efficient way of locating free parking spaces in the three-dimensional point clouds.

Problems solved by technology

However, it cannot be used when parking space markings are not visible.

Also, performance can be degraded by poor visual conditions such as stains, shadows or occlusion by adjacent vehicles.

However, this requires extra costs and space for the equipment because a stereo camera is not a general configuration of a rearview camera.

Sub-pixel accuracy is required when there are short -baselines between the two cameras, and point correspondences are difficult to find when there are wide baselines.

However, this approach cannot be used during the day due to the presence of sunlight.

However, these researchers did not present a free parking space detection method.

However, odometry information can be erroneous when road conditions are slippery due to rain or snow.

Images